Automated Spine Label Propagation Using 3D Volumetric Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging systems require manual and time-consuming labeling of spinal images across different views, leading to inaccuracies and inconsistencies, as existing automated solutions fail to provide continuous and accurate label propagation across orthogonal views.
Innovation Solution
A system and method for automated labeling of vertebral and inter-vertebral regions using a volumetric labeling approach, where labels are propagated across multiple views by aligning them with the spinal center line, allowing for efficient and accurate identification of image regions irrespective of the viewing plane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual labeling is used to label spinal images across different views, then labeling accuracy can be maintained, but the time required and manual effort increase significantly
Solution Approach 1:
The patent creates a 3D volumetric model of the spine from multiple 2D image views and automatically copies labels between views by projecting them through the volumetric model. When a label is placed in one view, the system propagates it to corresponding locations in other views using the 3D spatial relationships stored in the volumetric model, eliminating manual relabeling while maintaining consistency across all views.
Solution Approach 2:
The patent introduces a 3D volumetric model as an intermediary between 2D image views. This volumetric model stores spatial relationships and serves as a mediator that automatically transforms and propagates labels between different 2D projections, replacing the need for manual coordinate transformation and label placement across multiple views.
2Productivity
If automated label propagation is implemented between orthogonal views, then labeling speed increases, but label accuracy and consistency across views deteriorate
Solution Approach 1:
The patent transitions from 2D label propagation to 3D volumetric label propagation. By storing and processing label positions in a 3D volumetric space rather than propagating directly between 2D views, the system maintains accurate spatial relationships and enables precise label placement across all views while automating the process.
Solution Approach 2:
The system uses the 3D volumetric model to accurately copy and project labels between views. The volumetric model preserves precise spatial coordinates, allowing labels to be automatically placed at correct 3D locations when viewed from different 2D perspectives, maintaining both speed and accuracy.
3Ease of operation
If labels are propagated using a single point intersection method in orthogonal views, then the labeling process is simplified, but continuous label display across slices and accuracy deteriorate
Solution Approach 1:
The patent moves from 2D point intersection to 3D volumetric labeling. Instead of finding intersection points in 2D orthogonal views, the system creates and manipulates labels in a 3D volumetric space, enabling continuous label display across multiple slices and maintaining accuracy by preserving true 3D spatial relationships.
Solution Approach 2:
The volumetric model enables continuous label propagation across all slices and views simultaneously. As users navigate through different 2D views or 3D slices, labels continuously appear at correct positions based on the underlying 3D volumetric data, providing smooth and consistent label display throughout the entire volume.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A system and method of labeling orthogonal or otherwise spatially related image views and related images is provided. The present invention provides automated progression for the labeling of vertebral and intervertebral regions, propagation of labels between views and images within a series, centering of label regions relative to the spine, circular lists of predefined labels, and label displays for individual slices of an orthogonal or axial view as a user scrolls through the plurality of image slices of the given view. In a further aspect, the present invention provides automated labeling of vertebral and inter-vertebral regions when a user provides labels for the adjacent two inter-vertebral or vertebral regions.